British Heart Journal, I972, 34, 874-88I. Br Heart J: first published as 10.1136/hrt.34.9.874 on 1 September 1972. Downloaded from Duration of phases of left ventricular systole using indirect methods I: Normal subjects
J. Fabian,' E. J. Epstein, and N. Coulshed From the Liverpool Regional Cardiac Centre, Sefton General Hospital, Liverpool
The duration of the various phases of left ventricular systole has been measured in a series of So normal subjects using high speed simultaneous recordings of the electrocardiogram, the phono- cardiogram, the carotid pulse, and the apex cardiogram. The five phases of systole measured were: total electromechanical systole (Q-A2 interval); left ventricular ejection time (LVET); pre-ejection period (PEP); isovolumetric contraction time (IVCT); total mechanical systole (TMS). The Q-A2 interval, left ventricular ejection time, and total mechanical systole were linearly and inversely related to heart rate. The pre-ejection period and the isovolumetric contraction time were not significantly related to heart rate. The validity of the method has been substantiated by comparing simultaneous records ofdirect and indirect pulseforms in a small series ofpatients. The are and at heart rates showed no deviation findings reproducible serial daily studies different from copyright. the normal range. The results have been compared with those reported by various authors and they are in close agreement.
Methods used to assess cardiac function, 2) Left ventricular ejection time (LVET).
which depend on measurements ofintravascu- 3) Pre-ejection period (PEP). http://heart.bmj.com/ lar pressure and flow, are not readily per- 4) Isovolumetric contraction time (IVCT). formed in ambulant patients. The procedures 5) Total mechanical systole (TMS). are uncomfortable, carry some risk to the patient, and are not easily repeated. The results in normal subjects are presented The duration of the phases of left ventricu- to establish a normal range for comparison lar systole, which are potentially useful as with those in various cardiac disorders. In indices of left ventricular performance, can be addition, we have compared our findings with measured from accurately externally recorded those of other workers. on September 29, 2021 by guest. Protected pulse waves, and such externally recorded time intervals do not appear to differ signifi- cantly from those obtained by direct measure- Methods and subjects ment. Recent publications have described the Studies were performed on 5o normal healthy various techniques which can be used (Jezek, adults (35 men and I5 women) aged i8 tO 79 I963; Oreshkov, I965; Weissler, Harris, and years. Twenty-five were under 40 and 25 were Schoenfeld, i968; Spodick and over 40 years of age. The majority were members Kumar, of the hospital staff and the remainder were sur- i968a and b). gical patients admitted for relatively minor pro- This paper describes a method of recording cedures unrelated to the cardiovascular system. and measuring five important phases of left All subjects had a normal exercise tolerance, no ventricular systole, viz: cardiovascular abnormality on clinical examina- tion, a normal electrocardiogram, and were not i) The Q-A2 interval or total electromechani- on any form of medication. cal systole (TEMS). Simultaneous recordings were made of the electrocardiogram, the phonocardiogram, the Received 28 October 1971. apex cardiogram, and the right carotid arterial 'Present address: First Medical Clinic, Charles Uni- pulse using a multichannel oscilloscopic recorder versity, Vlasska 36, Prague x, Czechoslovakia. (Cambridge Instrument Company) and a paper Duration ofphases of left ventricular systole using indirect methods 875 Br Heart J: first published as 10.1136/hrt.34.9.874 on 1 September 1972. Downloaded from
i) The total period of electromechanical systole or the Q-A2 interval. Defined as the interval from the onset of the Q wave of the electrocardiogram
2-e L to the onset ofthe first high frequency component of the aortic component of the second sound. 2) The left ventricular ejection time measured from the beginning of the carotid upstroke to the nadir of the carotid incisura. 3) The interval from the beginning of the up- stroke of the apex cardiogram to the onset of the carotid upstroke. 4) The delay in transmission ofthe central arterial pulse to the carotid artery, or the pulse trans- mission time, measured from the first rapid vibration of the aortic component of the second sound to the nadir of the carotid incisura. ;ii All intervals were calculated as the mean of I1 ( measurements made on at least five cardiac cycles, each read to the nearest 5 msec. Heart rate was ,FIG. i Normal subject: simultaneous records calculated from the average RR interval of the of the apex cardiogram (ACG), the carotid cycles measured. pulse (CP), the phonocardiogram (PCG), and The following intervals were calculated from lead II of the electrocardiogram. A = Q-A2 the above measurements. interval; B = left ventricular ejection time i) Pre-ejection period This was calculated (LVET); C= onset of ACG upstroke (u) to by subtracting the left ventricular ejection time onset of CP upstroke (u); D = A2 to dicrotic from the total period of electromechanical systole notch interval or pulse transmission time (Q-A2 interval). PEP = (Q-A2)- (LVET). This (PTT); E= total mechanical systole (TMS); is the same as the interval from the onset of the copyright. A minus B =pre-ejection period (PEP); Q wave of the electrocardiogram to the onset C minus D = isovolumetric contraction time of the carotid upstroke corrected for the pulse transmission (IVCT); S1=first heart sound; A2, P2 = delay. aortic and pulmonary components of the second 2) Isovolumetric contraction time (IVCT) heart sound; 2L = second left intercostal space The interval from the beginning of the upstroke parasternally; MF= medium frequency. Time of the apex cardiogram (ACGu) to that of the are at msec klines 40 intervals. carotid pulse (CARu) corrected for the delay in http://heart.bmj.com/ transmission of the central arterial pulse to the carotid artery (PTT) (Oreshkov, I965; Spodick and Kumar, i968a). IVCT=ACGu to CARu- speed of IOO mm/sec (Fig. I). The recordings PTT. were performed in a similar fashion to the routine phonocardiographic practice of this Unit (Coul- 3) Total mechanical systole (TMS) The shed and Epstein, I963). Usually standard lead II sum of isovolumetric contraction time (IVCT) of the electrocardiogram was employed. A crystal and left ventricular ejection time (LVET). microphone to record the heart sounds was placed TMS=IVCT+LVET. This is the same as the parasternally in the second or third left interspace. interval from the beginning of the upstroke of the on September 29, 2021 by guest. Protected Occasionally two microphones were used in order apex cardiogram to the aortic component of the to identify the aortic and pulmonary components second heart sound (A2). of the second sound. The right carotid arterial The results on the 50 subjects were statistically pulse and apex cardiogram were recorded using evaluated (Croxton, 1959). Standard deviations funnel-shaped pick-ups with an internal diameter and regression equations were calculated to relate ofapproximately I *5 cm. The funnels were connec- heart rate to systolic time intervals. ted to identical piezoelectric crystal microphones In ii patients with various cardiac disorders a %by similar lengths of rigid plastic tubing less than Telco micromanometer was placed in the central 2 feet in length. There was no time lag between aorta and left ventricle during a diagnostic study. the two recording systems. Records were taken Simultaneous records were made of the carotid with the breath held in partial expiration and with pulse and the central aortic pressure curve in 7 the patient in a semi-left lateral position, in order patients and of the apex cardiogram and the left to bring the apex impulse against the chest wall. ventricular pressure curve in 7 patients. Three of Identification of the underlying zone of the ven- these patients had both aortic and then left ven- tricle was accomplished by recording a unipolar tricular pressures recorded with the carotid pulse -electrocardiogram lead at the site of the apex and the apex cardiogram, respectively. Experi- cardiogram. mental studies showed that there was no signifi- The following intervals were measured directly cant time lag between the Telco micromanometer from the tracing (Fig. i). and the external pulse microphone. 876 Fabian, Epstein, and Coulshed Br Heart J: first published as 10.1136/hrt.34.9.874 on 1 September 1972. Downloaded from
Results electrocardiogram, the phonocardiogram, the These are shown in Table and Fig. 2. carotid pulse, and the apex cardiogram. With a fast recording speed of IOO mm/sec, mea- i) Q-A2 interval or total electro- surements can be made within an accuracy of mechanical systole (TEMS) The mean about 5 msec. Braunwald, Sarnoff, and Stainsby (I958) Q-A2 interval was 382 ± 23 msec with a range of 330 to 440 msec. There was a close linear studied left ventricular ejection time in an iso- relation between the Q-A2 interval and heart lated dog heart preparation. They found that rate (Fig. 2). The regression equation was increasing the stroke volume at a fixed heart rate increased the ejection time, whereas in- Y= + 503-I *59 HR (r= -0 79; P < OOOI). creasing the heart rate with a fixed stroke 2) Left ventricular ejection time (LVET) volume shortened the ejection time. Wallace The mean LVET was 28I ±2I msec with a et al. (I963) studied in more detail the various range of 230 tO 334 msec. There was a close phases ofleft ventricular systole in dogs. They linear relation between the LVET and heart showed that increasing the stroke volume at rate (Fig. 2). The regression equation was a fixed heart rate prolonged the ejection time, Y = + 389-I14I HR (r = -078; P
TABLE I Systolic time intervals in .5o normal subjects Range Mean Standard Regression equation deviation Total electromechanical systole (Q-A2 interval) 330-440 382 + 23 503--I 59 HR SD= ± I4 r= - 079, P 350- 450 Q-A2 +503 -1-59 HR LVET= +389 -I141HR r -0-79 p4Xl0I r= -0-78 p<0001 SD= ± 14 SD- ± 13 u 300 - E . . ~250 300- ,20C0 - 0 50 60 70 80 90 100 50 60 76 80 90 10o Heart rate Heart rate PEP=126-0-34HR 150I r- -0.30 p<0-05 SD= i 12 0 u * .* E _ * * *- s *. j *' * : cin, 3' I: e.g. copyright. 50s ©P ~~. 50' 60 70 80 90 10o Heart rate Heart rate FIG. 2 (A) Relation of Q-A2 interval to heart rate in 50 normal subjects. In this and http://heart.bmj.com/ subsequent Figures the thick line is the regression line and the fine line shows ±2 standard deviations. (B) Relation of left ventricular ejection time (LVET) to heart rate in 50 normal subjects. (C) Relation ofpre-ejection period (PEP) to heart rate in 50 normal subjects. (D) Relation of total mechanical systole (TMS) to heart rate in so normal subjects. heart rate in normal subjects appears to be point. Simultaneous measurements of left mainly due to shortening of left ventricular ventricular ejection time were made using a on September 29, 2021 by guest. Protected ejection time, since the pre-ejection and the Telco micromanometer in the central aorta isovolumetric contraction time did not vary and a piezocrystal pulse microphone over the significantly with heart rate. right carotid artery. There was no significant The left ventricular ejection period is an difference between the direct and indirect important measurement of cardiac perform- measurements of ejection time (Table 2 and ance and occurs between the opening and Fig. 3), and carotid pulse tracings can there- closing of the aortic valve. It is measured from fore be used for measurement of left ventricu- the onset of pressure rise to the incisural notch lar ejection time. However, accurate measure- of the central aortic pressure pulse. Left ven- ment requires care in recording the carotid tricular ejection time can also be measured pulse wave in order to show a clear onset of from the indirect carotid pulse tracing, and in the carotid upstroke and a definite incisural 5 patients studied by Weissler, Peeler, and notch. Roehil (I96I), there was close agreement with The left ventricular ejection time in our the ejection time measured from a simul- series was 28I ±2I msec and was similar to taneously recorded central aortic pressure the findings of Jezek (I963) who found an pulse. ejection time of 274 ± 2imsec: of Spodick and We have studied 7 patients to validate this Kumar (I968a) who found a time of 292 ± I9 878 Fabian, Epstein, and Coulshed Br Heart J: first published as 10.1136/hrt.34.9.874 on 1 September 1972. Downloaded from TABLE 2 Comparison of apex cardiogram and carotid pulse with simultaneous records of left ventricular pressure and aortic pressure using TELCO intravascular micromanometer. (All measurements are given in msec and are the mean of 5 cardiac cycles.) Patient No. and diagnosis Onset apex Simultaneous Simultaneous cardiogram to onset of LV LVET* LVET Onset aortic A2 to carotid pressure aorta carotid pulse pressure to pulse dicrotic (TELCO) (TELCO) onset of notch carotid pulse i) Cardiomyopathy I7 229 227 29 25 2) LV aneurysm and VSD 22 212 2I3 29 29 3) MR I7 210 209 28 30 4) MR I9 - - 5) Cardiomyopathy 3I - 6) Congestive cardiomyopathy IS - 7) Congestive cardiomyopathy i- 8) MR and MS - 215 2I7 15 17 9) Mild MR 282 286 25 25 io) Constrictive pericarditis - 246 246 i8 I7 ii) LV aneurysm 221 220 I6 I6 Average 20 23I 23I 23 23 LVET =left ventricular ejection time. msec: and of Sawayama et al. (I969) who heart rate (r = - 0.30). The regression equa- found a time of 274 ± 2I msec. There was a tion was similar to that calculated by Weissler linear and inverse relation to heart rate (Fig. et al. (I968) in their series of normal subjects copyright. 2), and the regression equation was also (Fig. 2). similar to that found by Spodick and Kumar The preceding measurements have used (I968a) and Weissler et al. (I968). the electrocardiogram, the phonocardiogram, Weissler et al. (I96I) studied the relation and the carotid pulse wave. For indirect mea- between left ventricular ejection time, heart surement of the isovolumetric contraction rate, and stroke volume in normal resting time of the left ventricle a simultaneous apex subjects. They found that the ejection time cardiogram is also recorded. Spodick and http://heart.bmj.com/ decreased as heart rate increased and increased Kumar (I968b) have studied the use of the as stroke volume was augmented, both rela- simultaneous apex cardiogram and carotid tions being linear. In a clinical study, Harley, pulse wave to measure the isovolumetric Starmer, and Greenfield (I969) also showed contraction time of the left ventricle and con- that the duration of ejection was directly and sidered three definitions: firstly, the time in- linearly related to stroke volume and inversely terval from mitral valve closure to the onset related to heart rate. These findings are con- of left ventricular ejection; secondly, the time sistent with the experimental work carried out interval from the onset of left ventricular pres- on September 29, 2021 by guest. Protected in dogs by Braunwald et al. (I958) and by sure rise, which precedes mitral valve closure, Wallace et al. (I963). It is therefore clear that to the onset of ejection (Wiggers, I92I); and heart rate and stroke volume each have an thirdly, the time interval from the onset of independent influence on the duration of left left ventricular contraction or movement, ventricular ejection. which precedes the rise of pressure, to the The pre-ejection period is the time interval onset of ejection. This third definition is that between the onset of depolarization and the used by Spodick and Kumar (I968b). It has onset of left ventricular ejection. It includes been described previously by Oreshkov (I965) the time taken for the spread of the electrical and we have used it in this study. It entails the stimulus through the myocardium; the period use of the apex cardiogram, which is the only of left ventricular contraction before mitral simple clinical method available for detecting valve closure; and the period between mitral the earliest movement of the left ventricle. valve closure and the onset of ejection. It is At the onset of left ventricular systole the equal to the difference between the Q-A2 ventricular wall begins to move before the interval and the left ventricular ejection time. pressure in the left ventricle starts to rise. This In this series of 5o normal subjects, the pre- early motion of the left ventricular wall is ejection period showed a poor correlation with first detected by the apex cardiogram. Tafur, Duration ofphases of left ventricular systole using indirect methods 879 Br Heart J: first published as 10.1136/hrt.34.9.874 on 1 September 1972. Downloaded from findings differ from those of Willems, De Geest, and Kesteloot (1971) in dogs, and of Bush et al. (I970) in man, who failed to show any delay between the onset of the upstroke r iNTIA A01.flC KIG of the apex cardiogram and the onset of the rise of left ventricular pressure. 100 This definition of isovolumetric contraction time includes the total period of left ventricu- lar contraction before ejection, i.e. the period of rising left ventricular intramural tension NL 2LMF and the period of rising left ventricular intra- cavitary pressure up to the onset of aortic valve opening. It includes the preisometric contraction phase of Wiggers (I92I). Q- )OT1D PUL5E. The period of isovolumetric contraction ends with the onset of ejection. This is identi- 0 fied by recording the onset of the carotid pulse .6 H~~~~~~I FIG. 3 Simultaneous tracings of the central FI G. 4 Simultaneous tracings of the apex aortic pressure pulse recorded with an intra- cardiogram (ACG) (lower tracing) and of the cardiac micromanometer, and the indirect left ventricular pressure pulse recorded with an carotid pulse recorded with a piezocrystal intracardiac micromanometer (upper tracing) > transducer. The top tracing shows an intra- from a patient with idiopathic cardiomyopathy. aortic phonocardiogram (PCG) and the third The top tracing shows an intracardiac phono- copyright. tracing shows an external PCG recorded at cardiogram recordedfrom the left ventricle and the second left intercostal space parasternally the second tracing an external phonocardio- (2L). Left ventricular ejection time measured gram. A = the atrial displacement impulse of from both the aortic and carotid pulse waves the apex cardiogram; a = the left atrial pressure was 230 msec. B = onset of the aortic pressure pulse recordedfrom the left ventricle; I = onset pulse; C = onset of the indirect carotid pulse; of the QRS complex of the electrocardiogram; D = onset of the aortic component of the second 2 = onset of the apex cardiogram; 3 = onset of http://heart.bmj.com/ heart sound (A2); E= the nadir of the dicrotic the left ventricular pressure pulse. The apex notch (DN) of the indirect carotid pulse. The cardiogram clearly precedes the left ventricular intervals B to C and D to E both equal 30 pressure pulse by the time interval between 2 msec. and 3, which is IS msec. SI, S2, S3, S4 are the first, second, third, andfourth heart sounds. on September 29, 2021 by guest. Protected Ii 02 53 Cohen, and Levine (I964) first noted this difference in 2 patients with simultaneous L.VENT]CLE PCG 1\ 1!t I15 't records of the apex cardiogram and left ven- tricular pressure. The upstroke of the apex cardiogram preceded the onset of the rise in 100 left ventricular pressure by i8 milliseconds. PIG 2L 4F Inoue et al. (I970) found experimentally in dogs that the onset of the apex cardiogram preceded the rise of left ventricular pressure by I7 msec. In the 7 patients we have studied LEFT VENTfICLE with simultaneous records of the apex cardio- gram and left ventricular pressure, the up- AU, stroke of the apex cardiogram preceded the 8 rise of left ventricular pressure by an average of 20 msec (Fig. 4 and Table 2). This agrees with the findings of Tafur et al. (I964) and confirms their observations. However, our 88o Fabian, Epstein, and Coulshed Br Heart J: first published as 10.1136/hrt.34.9.874 on 1 September 1972. Downloaded from wave simultaneously with the apex cardio- This work was aided by a Research Grant from gram and subtracting the time taken for the the Medical Research Committee of the United and also a research grant pulse wave to travel from the central aorta to Liverpool Hospitals by from the Peel Medical Trust, for which we are the carotid artery (Fig. i). This time lag, the most grateful. pulse transmission time, appears to be the same at the onset as at the end of ejection (Harrison et al., I964), and equal to the time interval between the aortic second sound and the nadir ofthe carotid incisura. We have been References unable to find any data to substantiate this Braunwald, E., Sarnoff, S. J., and Stainsby, W. N. point and therefore we studied 7 patients using (1958). Determinants of duration and mean rate of simultaneous recordings of central aortic pres- ventricular ejection. Circulation Research, 6, 3I9. sure and the indirect carotid pulse. The delay Bush, C. A., Lewis, R. P., Leighton, R. F., Fontana, M. E., and Weissler, A. M. (1970). Verification of between the onset of the aortic pressure pulse systolic time intervals and true isovolumic contrac- and the right carotid pulse wave was on aver- tion time from the apex cardiogram by micromano- age 23 msec (range I9-29), (Table 2 and Fig. meter catheterization of the left ventricle and aorta. 3). The interval between the aortic second Circulation, 42, Suppl. III, I2I. Coulshed, N., and Epstein, E. J. (I963). The apex sound and the nadir of the carotid incisura in cardiogram: its normal features explained by those these patients was also on average 23 (range found in heart disease. British Heart Journal, 25, I6-30) msec. These two measurements were 697. therefore in close agreement, thus validating Croxton, F. E. (1959). Elementary Statistics. Dover Publications, New York. the use of the A2 - incisura interval as an Harley, A., Starmer, C. F., and Greenfield, J. C. accurate measure of the pulse transmission (I969). Pressure-flow studies in man. An evaluation time from the central aorta to the carotid of the duration of the phases of systole. Jrournal of artery. Clinical Investigation, 48, 895. Harrison, T. R., Dixon, K., Russell, R. O., Bidwai, The isovolumetric contraction time in the P. S., and Coleman, H. N. (I964). The relation of 50 normal subjects studied was 70 ± 9.5 msec age to the duration of contraction, ejection, and (range 5i-9o). This agrees with the findings of relaxation of the normal human heart. American copyright. Oreshkov (i965) who found it to be 67 ± I5 Heart_Journal, 67, I89. Inoue, K., Young, G. M., Grierson, A. L., Smulyan, msec (range 40-9o) and Spodick and Kumar H., and Eich, R. H. (I970). Isometric contraction (I968b) who found it to be 70-9 ± i5-8 msec period of the left ventricle in acute myocardial (range 40 to go). There was no significant infarction. Circulation, 42, 79. relation between the isovolumetric contraction Jezek, V. (I963). Clinical value of the polygraphic time and heart rate, and our confirm tracing in the study ofthe sequence of events during findings cardiac contraction. Cardiologia, 43, 298. http://heart.bmj.com/ those of Oreshkov (I965) who also found no Oreshkov, V. (I965). Indirect measurement of iso- relation to heart rate in a study of 40 normal volumetric contraction time on the basis of poly- subjects. This differs from the experimental graphic tracing. Cardiologia, 47, 3I5. et who Sawayama, T., Ochiai, M., Marumoto, S., Matsuura, studies of Wallace al. (I963) found T., and Niki, I. (I969). Influence of amyl nitrite that the isovolumetric contraction time in dogs inhalation on the systolic time intervals in normal was significantly shortened by increasing the subjects and in patients with ischemic heart disease. heart rate. Circulation, 40, 327. Serial daily studies of the phases of left ven- Spodick, D. H., and Kumar, S. (I968a). Left ventricu- lar ejection period. Measurement by atraumatic on September 29, 2021 by guest. Protected tricular systole in normal subjects, before and techniques: results in normal young men and com- after exercise, have shown no deviation of the parison of methods of calculation. American Heart measurements from the normal range with Journal, 76, 70. heart rates varying from 50 to I20 beats a Spodick, D. H., and Kumar, S. (I968b). Isovolumetric contraction period of the left ventricle. Results in a minute. Observations in the same subject are normal series and comparison of methods of calcu- reproducible from day to day. lation by atraumatic techniques. American Heart The methods described in this paper pro- 7ournal, 76, 498. vide a simple non-traumatic technique for Tafur, E., Cohen, L. S., and Levine, H. D. (I964). The normal apex cardiogram. Its temporal relation- measuring cardiac performance. Measure- ship to electrical, acoustic and mechanical cardiac ments can be repeated as often as required events. Circulation, 30, 38I. without any disturbance or risk to the patient. Toutouzas, P., Gupta, D., Samson, R., and Shilling- The normal measurements and their relation ford, J. (I969). Q-second sound interval in acute myocardial infarction. British Heart Journal, 31, to heart rate provide a basis for comparison 462. with the findings in cardiac disease. Wallace, A. G., Mitchell, J. H., Skinner, N. S., and Sarnoff, S. J. (I963). Duration of the phases of left ventricular systole. Circulation Research, 12, 6i I. We wish to thank Mrs. B. Lea, chief cardiological Weissler, A. M., Harris, W. S., and Schoenfeld, C. D. technician, and the members of her staff for in- (I968). Systolic time intervals in heart failure in valuable technical help. man. Circulation, 37, 149. Duration ofphases of left ventricular systole using indirect methods 88i Br Heart J: first published as 10.1136/hrt.34.9.874 on 1 September 1972. Downloaded from Weissler, A. M., Peeler, R. G., and Roehll, W. H. Willems, J. L., De Geest, H., and Kesteloot, H. (I97I). (I96I). Relationships between left ventricular ejec- On the value of apex cardiography for timing intra- tion time, stroke volume, and heart rate in normal cardiac events. American3Journal of Cardiology, 28, individuals and patients with cardiovascular disease. 59. American Heart Journal, 62, 367. Wiggers, C. J. (I92I). Studies on the consecutive phases of the cardiac cycle: II. Laws governing the Requests for reprints to Dr. E. J. Epstein, Liver- relative durations of ventricular systole and di- pool Regional Cardiac Centre, Sefton General astole. American Journal of Physiology, 56, 439. Hospital, Smithdown Road, Liverpool LI5 2HE. copyright. http://heart.bmj.com/ on September 29, 2021 by guest. Protected